In this study we examine the regression-based ratio-correlation method and suggest some new tools for assessing the magnitude and impact of coefficient instability on population estimation errors. We use a robust sample of 904 counties from 11 states and find that: (1) coefficient instability is not a universal source of error in regression models for population estimation and its impact is less than commonly assumed; (2) coefficient instability is not related to bias, but it does decrease precision and increase the allocation error of population estimates; and (3) unstable coefficients have the greatest impact on counties under 20,000 in population size. Our findings suggest that information about the conditions that affect coefficient instability and its impact on estimation error might lead to more targeted and efficient approaches for improving population estimates developed from regression models.
BryanT., Population estimation, in: Methods and Materials of Demography, Second Edition, J.S. Siegel and D.A. Swanson, eds, Amsterdam: Elsevier, 2004, pp. 523-560.
2.
ChowG.C., Tests of equality between sets of coefficients in two linear regressions, Econometrica28 (1960), 591-605.
3.
CrosettiA.H. and SchmittR.C., A method for estimating the inter-censal population of counties, Journal of the American Statistical Association51 (1956), 587-590.
4.
DoughertyC., Introduction to econometrics, Fourth edition. Oxford: Oxford University Press, 2011.
5.
DuncanO.D., CuzzortR.P. and DuncanB., Statistical geography: Problems in analyzing area data, Glencoe: Free Press, 1961.
6.
EricksenE.P., A method for combining sample survey data and symptomatic indicators to obtain population estimates for local areas, Demography10 (1973), 137-160.
7.
EricksenE.P., A regression method for estimating population changes for local areas, Journal of the American Statistical Association69 (1974), 867-875.
8.
FonsecaL. and TaymanJ., Postcensal estimates of household income distributions, Demography26 (1989), 149-160.
9.
GoldbergD., RaoV.R. and NamboodiriN.K., A test of the accuracy of ratio correlation population estimates, Land Economics40 (1964), 100-102.
10.
GoldsteinH., Mltilevel Statistical Models, (4th Edition), Chichester: John Wiley, 2011.
11.
GoodallC., M-estimators of location: An outline of the theory, in: Understanding robust and exploratory data analysis, D.C. Hoaglin, F. Mosteller and J.W. Tukey, eds, New York: John Wiley, 1983, pp. 339-403.
12.
HarperG., ColemanC. and DevineJ., Evaluation of 2000 subcounty population estimates [Online]. Available: http://www.census.gov/population/www/documentation/twps0070/twps0070.html, 2003.
13.
HeritierS., CantoniE., CoptS. and Victorua-FeserM.P., Robust methods in biostatistics, Chichester: John Wiley, 2009.
14.
HoaglinD.C., MostellerF. and TukeyJ.W., M-estimators of location: An outline of the theory, in: introduction to more refined estimators, D.C. Hoaglin, F. Mosteller, and J.W. Tukey, Eds., ed New York: John Wiley, 1983, pp. 283-296.
15.
HoqueN., Evaluation of small area population estimates produced by housing unit, ratio-correlation, and component method II compared to 2000 Census counts, Canadian Studies of Population39 (2010), 91-108.
16.
HosmerD.W., LemeshowS. and SturdivantR.X., Applied Logistic Regression, Hoboken: John Wiley, 2013.
17.
HuberP.J. and RonchettiE.M., Robust Statistics, (Second edition), New York: John Wiley, 2009.
18.
MandellM. and TaymanJ., Measuring temporal stability in regression models of population estimation, Demography19 (1982), 135-146.
19.
MasseyD.S. and DentonN.A., The dimensions of residential segregation, Social Forces67 (1988), 281-315.
20.
McKibbenJ.N. and SwansonD.A., Linking substance and practice: A case study of the relationship between socio-economic structure and population estimation, Journal of Economic and Social Measurement24 (1997), 135-147.
21.
NamboodiriN.K., On the ratio-correlation and related methods of subnational population estimation, Demography9 (1972), 443-453.
22.
NamboodiriN.K. and LaluN.M., The average of several regression estimates as an alternative to the multiple regression estimate in postcensal and intercensal estimates: A case study, Rural Sociology36 (1971), 187-194.
23.
O'HareW.P., Report on a multiple-regression method for making population estimates, Demography13 (1976), 369-379.
24.
O'HareW.P., Report on a multiple-regression method for making population estimates, Demography17 (1980), 341-343.
25.
PursellD.E., Improving population estimates with the use of dummy variables, Demography7 (1970), 87-91.
26.
RaoJ.N.K., Small Area Estimation, Hoboken: Wiley Interscience, 2003.
27.
RosenbergH., Improving current population estimates through stratification, Land Economics44 (1968), 331-338.
28.
SchmittR.C. and CrosettiA.H., Accuracy of the ratio-correlation method for estimating postcensal population, Land Economics30 (1954), 279-281.
29.
SchmittR.C. and GrierJ.M., A method for estimating the population of minor civil divisions, Rural Sociology31 (1966), 355-361.
30.
SerowW.J. and MartinJ.H., Estimating demographic characteristics using the ratio-correlation method, Demography15 (1978), 223-233.
31.
SiegelJ.S., Applied demography: Applications to business, government, law, and public policy. San Diego: Academic Press, 2002.
32.
SmithS.K. and CodyS., An evaluation of population estimates in Florida: April 2010, Population Research and Policy Review23 (2004), 1-24.
33.
SmithS.K. and CodyS., An evaluation of population estimates in Florida, Bureau of Economic and Business Research, Gainesville, 2011.
34.
SnowE.C., The application of the method of multiple correlation to the estimates of postcensal populations, Journal of the Royal Statistical Society74 (1911), 575-629.
35.
SparM.A. and MartinJ.H., Refinements to regression-based estimates of postcensal population characteristics, Public Data Use7 (1979), 16-22.
36.
SwansonD.A., An evaluation of ratio and difference regression methods for estimating small, highly concentrated populations, Review of Public Data Use6 (1978), 18-27.
37.
SwansonD.A., Improving accuracy in multiple regression estimates of county populations using principals of causal modeling, Demography17 (1980), 413-427.
38.
SwansonD.A., Allocation accuracy in population estimates: An overlooked criteria with fiscal implications, in: Small area population estimates and their accuracy, U.S. Bureau of the Census, Ed., ed Washington, D.C.: U.S. Bureau of the Census, 1981, pp. 13-21.
39.
SwansonD.A., Advancing methodological knowledge within state and local demography: A case study, Population Research and Policy Review23 (2004), 379-398.
40.
SwansonD.A. and PrevostR., A new technique for assessing error in ratio-correlation estimates of population: A preliminary note, Applied Demography1 (1985), 1-4.
41.
SwansonD.A. and TaymanJ., Subnational Population Estimates, Dordrecht: Springer, 2012.
42.
SwansonD.A. and TedrowL.M., Improving the measurement of temporal change in regression models used for county population estimates, Demography21 (1984), 373-381.
43.
TaymanJ. and SchaferE., The impact of coefficient drift and measurement error on the accuracy of ratio-correlation population estimates, The Review of Regional Studies15 (1985), 3-10.
44.
TaymanJ., SmithS.K. and RayerS., Evaluating population forecast accuracy: A regression approach using county data, Population Research and Policy Review30 (2011), 235-262.
45.
TaymanJ. and SwansonD.A., On the validity of the MAPE as a measure of population forecast accuracy, Population Research and Policy Review16 (1999), 299-322.
46.
WardM.D. and GleditschK.S., Spatial Regression Models, Thousand Oaks: Sage, 2008.
